5-HTTLPR polymorphism is associated with nostalgia proneness: The role of neuroticism

ABSTRACT

Nostalgia, a sentimental longing for the past, is a self-relevant and social emotion. Nostalgia proneness is associated with alleviation of distress or instability (e.g., neuroticism). Although nostalgia proneness is heritable, the specific molecular contributors to this heritability are unknown. We focused on a polymorphism in the promoter of the serotonin transporter gene (5-HTTLPR) as a possible biological basis of nostalgia proneness, because the serotonin system has been associated with sensitivity to negative experience. Participants (N = 397 adults) who had reported levels of nostalgia proneness were genotyped. A subsample also completed a measure of neuroticism. Participants with the 5-HTTLPR short allele were higher on nostalgia proneness than those without this allele. Neuroticism mediated the relation between 5-HTTLPR and nostalgia proneness. These findings enrich our understanding of the genetic and personality underpinnings of nostalgia.

KEYWORDS:

• Nostalgia
• neuroticism
• serotonin
• 5-HTTLPR
• genetics

Acknowledgments

We thank Mike Hanus for collecting the data.

Disclosure statement

No potential conflict of interest was reported by the authors.

Supplemental data for this article can be accessed here.

Notes

¹ Some participants received additional compensation in the form of money or course credit for completing a daily life experience study unrelated to the current research.

² A negative intraclass correlation indicates that the variability within dyads exceeds the variability across dyads.

³ Before HLM analyses, we tested for partner effect (i.e., whether one’s 5HTTLPR can predict her/his partner’s nostalgia or neuroticism). No partner effect emerged for nostalgia (B = −0.17, SE = 0.17, t[324.06] = −0.99, p = .324) or neuroticism (B = −0.14, SE = 0.17, t[89.00] = −0.80, p = .425).

⁴ We excluded the three singlets (dyads with data missing from one partner) from HLM analyses since they could not be included in the dyadic data structure.

⁵ It is common for researchers to differentiate dyad members on factors like gender and analyze them as distinguishable dyads, particularly in regards to data from heterosexual couples (Kenny & Ledermann, 2010). Given that not all couples in our study were heterosexual and researchers caution against overemphasizing gender differences in dyadic studies (Ackerman, Donnellan, & Kashy, 2011), we treated all couples in HLM as indistinguishable dyads. However, to examine whether it was necessary to treat couples as distinguishable, we excluded homosexual couples (N = 3) and tested for distinguishability with only heterosexual couples. Following Kenny et al. (2006), we added gender (1 = male, 2 = female) as a level-1 predictor into the HLM (Model 1, χ²[5] = 1146.21, Akaike’s Information Criterion [AIC; Akaike, 1987] = 1150.21). The gender effect was not significant (B = −0.25, SE = 0.15, t[163.12] = −1.68, p = .095), suggesting indistinguishability. The estimate for genetic effect (B = 0.36, SE = 0.18, t[321.53] = 2.00, p = .046) from Model 1 was almost the same as the one from the original HLM, which included all couples and treated them as indistinguishable. Furthermore, based on Model 1, we allowed for different variances across genders in the HLM (Model 2, χ²[6] = 1145.37, AIC = 1151.37). Then, we compared the two models and obtained a non-significant model change, ∆χ²(1) = −.842, p = .359. In addition, Model 1 without heterogeneity of variance across genders yielded a lower AIC value, which suggested better fit. Taken together, heterosexual dyad members were indistinguishable, or at least needless to be differentiated.

⁶ We also tested the significance of the genetic effect by including demographic variables (sample, age, and gender) as covariates in the HLM. We include the results in online Supplementary Materials.

⁷ As nostalgia is a way to counteract life stressors, we wondered whether it moderated the relation between 5HTTLPR and neuroticism. To test this alternative hypothesis, we ran an additional HLM with nostalgia, 5HTTLPR genotypes, and their interaction as level-1 predictors, and with neuroticism as the outcome. The interaction between nostalgia and 5HTTLPR was not significant (B = −0.05, SE = 0.12, t[86.14] = −0.46, p = .651). Nostalgia did not moderate the 5HTTLPR-neuroticism association.

⁸ For the convenience of comparison, d = 0.18 is equivalent to r = 0.09.

⁹ We know that memory, reward, and emotion regulation networks are also modulated by dopamine (Grace, 2016; Nobili et al., 2017). Therefore, it is possible that dopamine is involved in nostalgia. To explore this, we genotyped two polymorphisms in genes related to dopaminergic function: A VNTR in exon 3 of the dopamine receptor D4 gene (DRD4) and a functional SNP in the Catechol-O-methyltransferase gene (COMT, rs4680). Importantly, the DRD4 gene codes for a dopamine receptor (D4) primarily located in the prefrontal cortex (Lahti et al., 1998) and the COMT gene codes for the COMT enzyme, which is the primary means of terminating dopamine signaling in the prefrontal cortex (Käenmäki et al., 2010). However, neither of these genes was associated with nostalgia (ts < .9, ps > .3, rs < .05).

Additional information

Funding

This work was supported by the National Science Foundation [BCS1104118].